Computer Vision Engineer

Train and deploy object detection and classification models across diverse physical settings in real-time and cloud-based inference environments. About City Detect City Detect harnesses AI and computer vision to revolutionize urban management by using sensors to map the physical world and help communities respond to issues and track longitudinal changes. Founded in 2021, we recently raised a Series A and are actively partnering with major municipalities across Texas, California, Florida, and beyond to create cleaner, safer, and more livable cities. We value responsible practices, cutting-edge innovation, and collaborative community partnerships as we scale our platform and shape the future of urban living through technology. Description We're seeking an experienced Computer Vision Engineer to design, develop, and deploy object detection models that power City Detect's core products. You'll work hands-on with YOLO, transformer, and similar architectures to build and optimize detection pipelines that deliver reliable, real-world performance across heterogenous settings. What You'll Do Develop, train, and optimize YOLO-based object detection models for production deployment Build and maintain end-to-end computer vision pipelines in Python, from data preprocessing through inference and evaluation Collaborate with labeling and data teams to define annotation guidelines, ontologies, and quality standards that drive model performance Evaluate model performance using detection metrics (mAP, precision, recall, IoU) and drive continuous improvement Contribute to research and experimentation on new architectures, training techniques, and data augmentation strategies Requirements 3+ years of professional experience in computer vision, with a strong focus on object detection 2+ years of hands-on experience with YOLO architectures (YOLOv5, YOLOv8, YOLO11, or similar) Proficiency in Python for model development, training, evaluation, and deployment (2+ years) Strong understanding of object detection fundamentals: bounding boxes, anchor-based vs. anchor-free methods, NMS, IoU, mAP Experience with deep learning frameworks such as PyTorch or TensorFlow (2+ years) Proven experience of model ownership from data curation to production-ready and deployment Proven experience working with datasets with heterogeneous attributes (weather, geography, etc.) Nice to Have Proven experience fine-tuning transformer architectures (2+ years) SQL proficiency for querying detection results, labeling metrics, or model performance data Experience with roadside or infrastructure object detection (signs, signals, debris, pavement markings) Background in GovTech, public sector, or smart city projects Experience in automated driving, ADAS, or autonomous vehicle perception systems Familiarity with model-assisted labeling, active learning, or human-in-the-loop workflows Experience with edge deployment or model optimization (TensorRT, ONNX, quantization) Eligible for company equity incentive plan Fully remote position Unlimited PTO Health, vision, and dental insurance $100 monthly wellness stipend Bi-annual team retreat What to Expect in Our Hiring Process Our hiring process is designed to be thoughtful, efficient, and human. Candidates typically move through a short series of interviews over 2–3 weeks, starting with a 30-minute phone screening, followed by one or two technical conversations and a final interview with our CEO. We focus on cultural alignment, real-world technical understanding, and career goals—not coding puzzles or LeetCode-style tests. You'll hear back within 24 hours after each stage whenever possible. If an offer is extended, the role begins with a 30-day trial period where you'll take ownership of a meaningful project and receive clear, ongoing feedback to ensure mutual fit. Due to regulatory and operational requirements, we are currently only considering candidates based in the United States. J-18808-Ljbffr